uber_apr14= pd.read_csv('F:\Data Science projects\by_other\uber-pickups-in-new-york-city/uber-raw-data-apr14.csv',encoding='utf-8')

files=os.listdir(r'F:\Data Science projects\by_other\uber-pickups-in-new-york-city')[-7:]

files

files.remove('uber-raw-data-janjune-15.csv')

files

path=r'F:\Data Science projects\by_other\uber-pickups-in-new-york-city'

​

#blank dataframe

final=pd.DataFrame()

​

for file in files:

    df=pd.read_csv(path+"/"+file,encoding='utf-8')

    final=pd.concat([df,final])

final.shape

df=final.copy()

df.head()

df.shape

df.dtypes

​

df['Date/Time'] = pd.to_datetime(df['Date/Time'], format="%m/%d/%Y %H:%M:%S")

​

df.dtypes

​

df['weekday']=df['Date/Time'].dt.day_name()

df['day']=df['Date/Time'].dt.day

df['minute']=df['Date/Time'].dt.minute

df['month']=df['Date/Time'].dt.month

df['hour']=df['Date/Time'].dt.hour

df.dtypes

​

df.head()

​

df['Base'].unique()

df['day'].unique()

df['weekday'].unique()

import plotly.express as px

px.bar(x=df['weekday'].value_counts().index,

      y=df['weekday'].value_counts().values

      )

​

plt.hist(df['hour'])

​

for i,month in enumerate(df['month'].unique()):

    print(month)

​

plt.figure(figsize=(40,20))

for i,month in enumerate(df['month'].unique()):

    plt.subplot(3,2,i+1)

    df[df['month']==month]['hour'].hist()

​

​

​

for i in df['month'].unique():

    plt.figure(figsize=(5,3))

    df[df['month']==i]['hour'].hist()

​

import chart_studio.plotly as py

import plotly.graph_objs as go

from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot

trace1 = go.Bar( 

        x = df.groupby('month')['hour'].sum().index,

        y = df.groupby('month')['hour'].sum(),

        name= 'Priority')

iplot([trace1])

plt.figure(figsize=(10,6))

plt.hist(df['day'], bins=30, rwidth=.8, range=(0.5, 30.5))

plt.xlabel('date of the month')

plt.ylabel('Total Journeys')

plt.title('Journeys by Month Day')

​

​

plt.figure(figsize=(20,8))

for i,month in enumerate(df['month'].unique(),1):

    plt.subplot(3,2,i)

    df_out=df[df['month']==month]

    plt.hist(df_out['day'])

    plt.xlabel('days in month'.format(i))

    plt.ylabel('total rides')

​

sns.set_style(style='whitegrid')

sns.pointplot(x="hour",y="Lat",data=lat_df)

ax=sns.pointplot(x="hour",y="Lat", hue="weekday",data=df)

ax.set_title('hoursoffday vs latiitide of passenger')

df.head()

df['Base'].head()

df.groupby(['Base','month'])['Date/Time'].count()

base=df.groupby(['Base','month'])['Date/Time'].count().reset_index()

base

plt.figure(figsize=(10,6))

sns.lineplot(x='month',y='Date/Time',hue='Base',data=base)

​

def count_rows(rows):

    return len(rows)

by_cross = df.groupby(['weekday','hour']).apply(count_rows)

by_cross

pivot=by_cross.unstack()

pivot

plt.figure(figsize=(10,6))

sns.heatmap(pivot, annot=False)

df.head()

def heatmap(col1,col2):

    by_cross = df.groupby([col1,col2]).apply(lambda x:len(x))

    pivot=by_cross.unstack()

    plt.figure(figsize=(10,6))

    return sns.heatmap(pivot,annot=False)

## validating above Analysis through Heatmap

heatmap('day','hour')

heatmap('day','month')

df[df['month']==4]

​

heatmap('weekday','month')

​

plt.figure(figsize=(10,6))

​

plt.plot(df['Lon'], df['Lat'],'r+', ms=0.5)

plt.xlim(-74.2, -73.7)

plt.ylim(40.6,41)

​

df.head()

df_out=df[df['weekday']=='Sunday']

df_out.head()

df_out.groupby(['Lat','Lon'])['weekday'].count().reset_index()

from folium.plugins import HeatMap

import folium

from folium.plugins import HeatMap

basemap=folium.Map()

HeatMap(df_out.groupby(['Lat','Lon'])['weekday'].count().reset_index(),zoom=20,radius=15).add_to(basemap)

basemap

​

def plot(df,day):

    df_out=df[df['weekday']==day]

    df_out.groupby(['Lat','Lon'])['weekday'].count().reset_index()

    HeatMap(df_out.groupby(['Lat','Lon'])['weekday'].count().reset_index(),zoom=20,radius=15).add_to(basemap)

    return basemap

​

plot(df,'Sunday')

​

​

uber_15 = pd.read_csv(r'F:\Data Science projects\by_other\uber-pickups-in-new-york-city/uber-raw-data-janjune-15.csv',encoding='utf-8')

uber_15.head()

uber_15.shape

​

#Checking the minimum date in the uber_15

uber_15['Pickup_date'].min()

#Checking the maximum date in the uber_15

uber_15['Pickup_date'].max()

uber_15['Pickup_date'] =  pd.to_datetime(uber_15['Pickup_date'], format='%Y-%m-%d %H:%M:%S')

​

uber_15['weekday']=uber_15['Pickup_date'].dt.day_name()

uber_15['day']=uber_15['Pickup_date'].dt.day

uber_15['minute']=uber_15['Pickup_date'].dt.minute

uber_15['month']=uber_15['Pickup_date'].dt.month

uber_15['hour']=uber_15['Pickup_date'].dt.hour

uber_15.head()

px.bar(x=uber_15['month'].value_counts().index,

           y=uber_15['month'].value_counts().values)

​

ax=sns.countplot(uber_15['hour'])

ax.yaxis.set_major_formatter(tick.FormatStrFormatter('%.0f'))

​

uber_15.groupby(['weekday', 'hour'])['Pickup_date'].count()

uber_15.groupby(['weekday', 'hour'])['Pickup_date'].count().reset_index()

summary=uber_15.groupby(['weekday', 'hour'])['Pickup_date'].count().reset_index()

summary=summary.rename(columns = {'Pickup_date':'Counts'})

summary

plt.figure(figsize=(10,6))

sns.pointplot(x="hour", y="Counts", hue="weekday", data=summary)

​

​

uber_foil=pd.read_csv(r'F:\Data Science projects\by_other\uber-pickups-in-new-york-city/Uber-Jan-Feb-FOIL.csv')

uber_foil.head()

uber_foil['dispatching_base_number'].unique()

sns.boxplot(x = 'dispatching_base_number', y = 'active_vehicles', data = uber_foil)